MULTILINE, ^, and $ work based on the platform, but the files being
parsed are from other platforms. As a result, the current code for
extracting the actual_results.json version will not find it in
logfiles produced on Windows when run on Mac.
The code for extracting the exact actual_results.json file to use
from a logfile is itself something of a hack, as this information
should be provided ina more structured manner. This proposed
method of finding the exact file is no worse than the old one, and
in cases like above, better.
Review URL: https://codereview.chromium.org/789253002
This allows us to control SK_DYNAMIC_ANNOTATIONS_ENABLED by Sk*Config.h files.
This is a no-op today, because we control it from the compiler command line.
BUG=430815
Review URL: https://codereview.chromium.org/787003003
This prevents races when calling fPtsToUnit.getType() from multiple threads.
This introduces a small amount of redundant code in SkTwoPointRadialGradient,
but it's probably optimized together into no extra run-time work.
BUG=437511
Review URL: https://codereview.chromium.org/793763003
Previously, a function was called using dlsym in skia_launcher.
Add a static initializer that changes the setting, and include that for
the tools we automate for testing.
Also only do va_copy if we actually use it.
BUG=skia:2454
Review URL: https://codereview.chromium.org/753543003
This includes:
-Having an actual XP stage at the end of the gl pipeline.
-All Blending work is handled by XP until actually setting GL blend states
-GLPrograms test to test XP
BUG=skia:
Review URL: https://codereview.chromium.org/764643004
With this CL, related nanobench can be improved for 565 config.
bitmap_BGRA_8888_update_scale_bilerp 76.1us -> 46.7us 0.61x
bitmap_BGRA_8888_scale_bilerp 78.7us -> 47us 0.6x
bitmap_BGRA_8888_update_volatile_scale_bilerp 82.7us -> 46.9us 0.57x
BUG=skia:
Review URL: https://codereview.chromium.org/788853002
GrGlTextureRenderTarget inherits virtually from both GrGlRenderTarget and
GrGLTexture, which both have a 'wrap' flag. The passed in wrap setting could
be different for the two base classes, but since it's virtually inherited,
they share the same flag, so they're either both on, or both off.
As a result, we fail to delete the frambuffer.
To fix this, we now keep a separate isWrapped flag for GrGlRenderTarget.
BUG=437998
Review URL: https://codereview.chromium.org/791493003
In addition, NVPR makes this very complicated, and I haven't quite figured out a good way to handle it, so for now color and coverage DO live on optstate, but I will figure out some way to refactor that in future CLs.
BUG=skia:
Review URL: https://codereview.chromium.org/783763002
Nexus 6 appears to require a sized internal format for A8 textures, much
like other newer mobile devices. Changed to use sized format for A8
textures in general with ES 3.0.
Review URL: https://codereview.chromium.org/783523003
It is desirable that, when layer hoisting is disabled, the MPD and non-MPD timings be
roughly the same. Unfortunately, using a separate canvas for each tile (a requirement
for MPD) introduces its own discrepancy into the timing. Using a separate canvas for
each tile doesn't seem to make a difference for 8888 (see the non-MPD 8888 column below)
but slows down GPU rendering (see the non-MPD GPU column below). Since this is how
Chromium renders I propose switching to this regimen (even though it is "slowing down"
GPU rendering).
nanobench mean times (ms) with layer hoisting disabled (for desk_amazon.skp)
8888
MPD non-MPD
1 canvas (old-style) 0.628 1.71
separate (new-style) 0.795 1.63
GPU
MPD non-MPD
1 canvas (old-style) 2.34 1.69
separate (new-style) 2.32 2.66
Review URL: https://codereview.chromium.org/779643002
